PhD position "The Cognitive and Interactional Infrastructure of Adaptive Language Use", Radboud University (the Netherlands)

Like many other core human abilities, communication appears fairly straightforward, until we try to build artificial cognitive agents that can communicate effectively. Human language use is marked by its pragmatic flexibility, building on shared knowledge (a.k.a. common ground). In everyday situations, people quickly build common ground using pragmatic inferences. How they exactly do that turns out to be remarkably difficult to explain. For instance, state-of-the-art computational explanations either make unrealistic assumptions or fail to scale beyond toy domains to situations of real-world complexity. This poses a scientific puzzle. Together with a highly interdisciplinary team, you will study the formal theoretical aspects of this scientific puzzle and try to identify conditions under which it can and cannot be computationally solved efficiently and effectively. To this end, you will use a combination of formal modelling, computational complexity analysis and agent-based simulations.